Efficient Photocatalytic Reduction of Hexavalent Chromium by NiCo₂S₄/BiOBr Heterogeneous Photocatalysts

Abstract

For typical Cr(VI)-containing industrial wastewater, more efficient water treatment technologies need to be used to ensure that Cr(VI) concentrations are reduced to safe levels before discharge. Photocatalytic technology is highly efficient, environmentally friendly, and has been extensively used to address this demand. Herein, heterogeneous NiCo₂S₄/BiOBr photocatalysts with different ratios were prepared using a solvothermal method. When compared with pure NiCo₂S₄ and BiOBr, the NiCo₂S₄/BiOBr-30 had significantly increased adsorption capacity and visible-light-driven photocatalytic reduction activity for Cr(VI) removal. The improved adsorption performance of the NiCo₂S₄/BiOBr-30 was mainly due to its increased specific surface area, and the enhanced photocatalytic performance of the NiCo₂S₄/BiOBr-30 could be attributed to the improved separation and transfer of photogenerated carriers at the interface. Lastly, a possible enhanced photocatalytic Cr(VI) reduction mechanism of the NiCo₂S₄/BiOBr heterostructure was developed.

1. Introduction

Chromium (Cr) can be used as an industrial coolant and cooling tower water additive; thus, the wastewater from metal plating, leather tanning, and manufacturing electroplating companies contains high levels of Cr [1,2]. The primary oxidation states of Cr in wastewater are Cr(III) and Cr(VI). Cr(VI) is more toxic than Cr(III) and poses a substantial health threat to humans and wildlife, where it is known to cause lung and digestive cancer, as well as upper abdominal pain, nausea, vomiting, and severe diarrhea and bleeding symptoms [3,4,5,6]. Due to the extremely serious impact of Cr(VI) on the environment and human health, almost all countries and regions have established strict standards for chromium discharge concentrations in industrial wastewater [7]. The World Health Organization (WHO) has recommended that the maximum Cr(VI) level in drinking water be less than 0.05 mg/L and the limits for industrial effluents must be less than 0.10 mg/L [8]. Therefore, the wastewater must be treated to remove Cr(VI) ions before being discharged into our environment to meet stringent environmental quality standards [9,10,11,12,13,14,15].

The concentration of Cr(VI) in wastewater from industrial processes can range from a few milligrams per liter (mg/L) to hundreds or even thousands of mg/L. Therefore, appropriate treatments must be implemented in industry for typical Cr(VI)-containing industrial wastewater [16,17,18,19]. Numerous techniques, such as adsorption, ion exchange, chemical reduction, electrochemical processes, and photocatalytic reduction, were investigated for Cr(VI)-containing wastewater [20,21,22,23]. Adsorption is regarded as a viable and advantageous approach among the current techniques [24,25,26]. However, because actual wastewater is complex and has a highly variable pH and coexisting ions, the practical application of adsorption for the removal of Cr(VI) from contaminated water is still limited [27,28,29,30]. When combined with photocatalysis (a highly efficient and environmentally friendly technology), the adsorbed Cr(VI) can be reduced and removed to realize the adsorbent reactivation, which can highly increase the removal performances of Cr(VI) [31,32]. Also, photocatalysis is a heterogeneous reaction, a liquid–solid reaction here for the Cr(VI) removal, where the adsorption process is usually the first stage of the photocatalytic reaction, and the catalyst’s adsorption capacity has a great influence on the efficiency of the photocatalytic reaction [33,34]. Strong adsorption can bring more Cr(VI) ions close to the active sites, increasing the chance of photocatalytic reduction [35]. Adsorbed Cr(VI) ions have a higher propensity to accept electrons for the photocatalytic reduction.

NiCo₂S₄, a transition metal sulfide with a variety of exceptional properties, has shown great photocatalytic performance for particular applications thanks to its adaptability and tunability [36,37,38]. The majority of the visible spectrum can be efficiently absorbed by NiCo₂S₄ due to its moderate bandgap energy level, which supplies more energy for photocatalytic reactions [39]. Furthermore, NiCo₂S₄ has good tunability, meaning that its characteristics can be tuned by doping, composition, or surface modification [40,41,42]. For example, Jin et al. modified NiCo₂S₄ using multi-walled carbon nanotubes (MWCNTs), which can effectively raise the interfacial charge transfer rate and improve NiCo2S4’s photocatalytic activity for Cr(VI) reduction [43]. Heterojunctions are composite structures that create distinct interfaces between two or more distinct semiconductor materials [44]. Xiong et al. and Wu et. al. reported the construction of NiCo₂S₄/ZnIn₂S₄ hybrids with tight interfacial contact for an enhanced photocatalytic reduction performance in H₂ production [38,45]. It is possible to effectively transfer and separate photogenerated carriers in heterogeneous photocatalysts, which may lead to more electrons participating in the reduction process and converting Cr(VI) into Cr(III). Among the various recently developed photocatalysts, bismuth oxybromide (BiOBr) belongs to the BiOX (X = F, Cl, Br, I) family, a tetragonal [Bi₂O₂X₂] layered structure, where the c-axis consists of [Bi₂O₂] slabs doubly interleaved by slabs of Br atoms [46,47]. The Bi center is surrounded by four strong covalently bonded O atoms and four weak interlayered van der Waals-interacted halogen atoms. Due to such strong intralayer and weak interlayer interactions of BiOX, the induced internal electric field (IEF) in the BiOX-based heterogeneous photocatalysts can play a vital role in inhibiting photogenerated charge carriers’ recombination during photocatalytic Cr(VI) reduction [31,48,49].

Herein, we report on heterogeneous NiCo₂S₄/BiOBr photocatalysts created via the in situ growth of BiOBr on a NiCo₂S₄ surface, where the BiOBr coatings improved the stability of NiCo₂S₄ catalysts during the Cr(VI) removal. NiCo₂S₄/BiOBr photocatalysts were characterized by UV–Vis DRS, EIS, XRD, SEM, TEM, XPS, and BET. It was expected that the construction of heterojunctions would improve the adsorption and photocatalytic reduction performances in the treatment of Cr(VI)-containing wastewater. The effect of the mass ratios of NiCo₂S₄ and BiOBr on the Cr(VI) removal performances were investigated; the effects of pH and co-existing ions during the photocatalytic reduction of Cr(VI) were also investigated. Among the composite samples, the NiCo₂S₄/BiOBr-30 composite exhibited the best performance, with a removal efficiency of 96.67% for Cr (VI) in 60 min, and demonstrated good cycling stability, which remained above 75% after four cycles. Such an improved photocatalytic performance of NiCo₂S₄/BiOBr-30 was mainly due to the stronger Cr(VI) adsorption and active sites that resulted from its increased specific surface area. Also, the inherent electric field of the NiCo₂S₄/BiOBr heterostructure improved the separation and transfer of photogenerated carriers, which were beneficial to the photocatalytic Cr(VI) reduction. The band structure of the heterogeneous photocatalyst was also determined to investigate the possible enhanced photocatalytic Cr(VI) reduction mechanism.

2. Materials and Methods

2.1. Preparation

All reagents used here were analytical-grade purity and purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China).

Synthesis of NiCo₂S₄: The NiCo₂S₄ was synthesized via a two-step solvothermal process. In the first step, 0.40 g of polyvinyl pyrrolidone K30 (PVP K30) was dispersed by ultrasonication into a 60 mL aqueous solution that contained 30 mL of ethanol, 0.2179 g of Ni(NO₃)₂∙6H₂O, and 0.4362 g of Co(NO₃)₂∙6H₂O, and then 0.45 g of CO(NH₂)₂ was added to the above solution with continuous stirring. The dissolved solution was transferred to a 100 mL Teflon-lined stainless-steel autoclave and maintained at 130 °C for 12 h. The resulting precipitate was washed and dried overnight at 70 °C. In the second step, 0.2 g of the as-obtained precipitate and 0.6 g of Na₂S∙9H₂O were added into 60 mL H₂O by ultrasonication and continuous stirring, and then the dissolved solution was transferred to a 100 mL Teflon-lined stainless-steel autoclave and kept at 180 °C for 12 h. The precipitate was collected and washed thoroughly with deionized water and ethanol after it cooled down to room temperature. The NiCo₂S₄ was obtained after drying at 70 °C overnight.

Synthesis of BiOBr: 0.0831 g of Bi(NO₃)₃·5H₂O and 0.0204 g of KBr were added to a 60 mL solution that contained 30 mL H₂O and 30 mL C₂H₆O₂ with continuous stirring. Then, the mixed solution was transferred to the 100 mL Teflon-lined stainless-steel autoclave and maintained at 160 °C for 12 h. The precipitate was collected and washed thoroughly with deionized water and ethanol after it cooled down to room temperature. The BiOBr was obtained after it dried at 70 °C overnight and the yield of BiOBr was about 0.0366 g.

Synthesis of NiCo₂S₄/BiOBr: Different amounts of NiCo₂S₄ (0.366 g, 0.183 g, 0.1220 g, 0.0915 g, and 0.0732 g) were dispersed in a 60 mL solution that contained 30 mL H₂O and 30 mL C₂H₆O₂, and then 0.0831 g of Bi(NO₃)₃·5H₂O and 0.0204 g of KBr were added to the suspensions with continuous stirring. Then, the mixed solution was transferred to the 100 mL Teflon-lined stainless-steel autoclave and maintained at 160 °C for 12 h. The precipitate was collected and washed thoroughly with deionized water and ethanol after it cooled down to room temperature. The NiCo₂S₄/BiOBr composite samples were obtained after it dried at 70 °C overnight. Since the mass ratios of BiOBr to NiCo₂S₄ were 10:100, 20:100, 30:100, 40:100, and 50:100, the composite samples were defined as NiCo₂S₄/BiOBr-10, NiCo₂S₄/BiOBr-20, NiCo₂S₄/BiOBr-30, NiCo₂S₄/BiOBr-40, and NiCo₂S₄/BiOBr-50, respectively.

2.2. Characterizations

The phase of the obtained products was determined by X-ray diffraction spectroscopy (XRD, Bruker D8 Advance, Karlsruhe, Germany). The morpholo, gies were measured using a scanning electron microscope (SEM, FEI Quanta250 FEG, Hillsboro, OR, USA). The microstructure and elemental distribution of the samples were determined by mapping profiles using a transmission electron microscope (TEM, JEOL JEM-F200, Tokyo, Japan) and energy-dispersive X-ray spectroscopy (EDS). The surface composition and elemental chemical state of the products were analyzed using an X-ray photoelectron spectrometer (XPS, Thermo Scientific K-Alpha, Hillsboro, OR, USA) equipped with a monochromatic Al Kα X-ray source (1486.6 eV), and XPSPEAK41 software was used for the peak fitting. Nitrogen adsorption–desorption isotherms and pore size distribution curves of the products were analyzed using a nitrogen adsorption device (JW-BK200B, Beijing JWGB Sci&Tech Co., Ltd., Beijing, China), and the specific surface area was calculated using Brunauer–Emmett–Teller (BET) methods. The UV–Vis diffuse reflectance (DRS) spectra of the products were obtained using a UV–Vis spectrophotometer (UV–Vis, Shimadzu UV-2600, Kyoto, Japan).

2.3. Photocatalytic Performances

The photocatalytic reduction activity of Cr(VI) in the presence of NiCo₂S₄/BiOBr catalysts was evaluated using an aqueous K₂Cr₂O₇ solution in a 60 mL quartz glass reactor. A total of 10 mg of the synthesized products was added to 30 mL of 100 ppm Cr(VI) and ultrasonically dispersed for 2 min. Then, the solution was continuously stirred for 30 min in the dark to reach the adsorption–desorption equilibrium before the photocatalysis process. After this, the mixed solution was exposed under a 300 W Xe lamp with a VisREF (300–780 nm) filter (MC-PF300C, Beijing Merry Change Technology Co., Ltd., Beijing, China) with continuous stirring; the active light intensity was about 1000 mW/cm². Every 10 min, 1 mL of the suspension was sampled and filtered through a nylon membrane filter with 0.22 μm pores. The concentration of Cr(VI) was mainly determined by a colorimetric diphenylcarbohydrazide (DPC) method by calculating the absorbance of the pink complex at 542 nm. The removal rate of Cr(VI) at time t was calculated according to Equation (1):

$$\mathsf{\eta }={\displaystyle \frac{(C_0-C_t)}{C_0}}\times 100\%$$

(1)

where η—the removal rate of Cr(VI) solution at time t, C₀—the initial concentration of the Cr(VI) solution, and C_t—the concentration of Cr(VI) in solution at time t.

2.4. Electrochemical Measurements

The electrochemical workstation (CHI660E, Shanghai Chinstruments, Shanghai, China) was used for all electrochemical measurements. An Ag/AgCl electrode served as the reference electrode, a Pt foil as the counter electrode, and a photocatalyst film (1.5 cm × 1.5 cm) on an FTO glass sheet (1.5 cm × 2.5 cm) as the working electrode in a conventional three-electrode system with a 0.1 M Na₂SO₄ solution as the electrolyte solution. A total of 10 mg of the catalyst was dispersed into 0.5 mL of ethanol in a 1 mL centrifuge tube by ultrasonication, and 10 μL of Nafion was added into the solution by further ultrasonication. Then, a 50 μL mixed solution was dropped onto a clean conductive FTO glass using a pipette. After heating and evaporation, the FTO glass sheets with photocatalyst film were used as a working electrode for the subsequent electrochemical measurement. The photocurrent generated from the working electrode in the light-irradiated aqueous electrolyte solution was maintained at +0.6 V Ag/AgCl under continuous Ar purging. Electrochemical impedance plot (EIS) measurements were conducted with a constant DC voltage at +0.6 V (vs. Ag/AgCl) under continuous Ar purging in the frequency range of 0.05 to 10⁵ Hz.

2.5. Simulated Calculation Details

The first-principles DFT calculations were carried out within the framework of density generalization theory using the Vienna ab initio calculation simulation package’s implementation of the projection augmentation plane wave method. For the exchange-correlation potential, Perdew, Burke, and Ernzerhof’s generalized gradient approximation was selected. The DFT-D3 method describes van der Waals interactions at long ranges. A cutoff energy of 400 eV was set for the plane wave. The Kohn–Sham equation was solved iteratively with the energy criterion set at 10⁻⁵ eV. Utilizing the Monkhorst–Pack scheme, Brillouin zone integrations were approximated. The maximum stress on each atom was kept within 0.02 eV/Å when optimizing the equilibrium geometries and lattice constants.

3. Results and Discussion

3.1. Structural Characterizations

The crystal phases of the pure NiCo₂S₄, BiOBr, and NiCo₂S₄/BiOBr composite products were examined using an XRD spectroscope. As shown in Figure 1, the characteristic peaks of the NiCo₂S₄ were located at 16.3°, 26.8°, 31.6°, 38.3°, 50.5°, and 55.3°, which corresponded to the (111), (220), (311), (400), (511), and (440) crystal planes of the NiCo₂S₄ (JCPDS No. 20-0782). The characteristic peaks of the BiOBr sample were located at 21.9°, 25.2°, 31.7°, 32.2°, 39.4°, 46.2°, 50.7°, 57.1°, 67.4°, 71.0°, and 76.7°, which corresponded to the (001), (002), (101), (102), (110), (112), (200), (104), (212), (220), (214), and (310) crystal planes of BiOBr (JCPDS No. 09-0393) [50]. The distinctive peaks of the NiCo₂S₄ and BiOBr were observed in the XRD pattern of the NiCo₂S₄/BiOBr composite, and the (102), (110), and (200) crystal planes of the BiOBr were responsible for the additional peaks at 31.7°, 32.2°, and 46.2°, which confirmed the in situ growth of the BiOBr on the NiCo₂S₄ surface.

The surface morphologies of NiCo₂S₄, BiOBr, and NiCo₂S₄/BiOBr composite products were examined using TEM and SEM. According to Figure 2, the shape of the NiCo₂S₄ nanoparticles was irregular, and the agglomeration microstructure was obviously observed. The BiOBr consisted of many nanospheres with a diameter range from 1 to 5 μm, and the aggregation of nanorod-like structures arranged these nanospheres into comparatively rough surfaces. For the NiCo₂S₄/BiOBr-30 composite sample, there was reduced agglomerations of the NiCo₂S₄ nanoparticles and close contact between the NiCo₂S₄ and BiOBr. Accordingly, the dispersion of the NiCo₂S₄ nanoparticles was facilitated by the composite of NiCo₂S₄ and BiOBr. The nitrogen adsorption–desorption isotherms for the NiCo₂S₄ and NiCo₂S₄/BiOBr composite samples are shown in Figure 2. In general, the specific surface area of the nanostructures was determined, and this surface area had a significant influence on the adsorption and reactions of pollutants. The isotherm types of the prepared samples all exhibited type-II isotherms with obvious H3 hysteresis loops, indicating that the NiCo₂S₄ and NiCo₂S₄/BiOBr-30 composite samples contained significant mesoporous structures [51]. The nitrogen adsorption–desorption isotherms for the other NiCo₂S₄/BiOBr-30 composites are demonstrated in Figures S1 and S2. According to these adsorption-desorption isotherm data, the samples’ specific surface area, pore volume, and pore size results are displayed in Table S1. The specific surface areas of the NiCo₂S₄, BiOBr, and NiCo₂S₄/BiOBr-30 were 14 m²/g, 21 m²/g, and 30 m²/g, respectively. Compared with the pure NiCo₂S₄ and BiOBr, the specific surface area of the NiCo₂S₄/BiOBr composite samples increased slightly, indicating that the BiOBr growth could increase the specific surface area. The increased specific surface area could partially provide more absorption and reaction sites for the Cr(VI) ions, and thus, improved the contact between catalyst and reactant [48]. This increased the adsorption capacity and the photocatalytic activity of the catalyst [52,53,54].

Using a relatively simplified model, density functional theory (DFT) calculations were performed in Figure 3. The electronic structural properties of the interface were first investigated for various catalysts with different densities of states (DOSs). The DOS diagrams of BiOBr, NiCo₂S₄, and BiOBr/NiCo₂S₄ are shown in Figure 3a–c. The dimensional information of different hyper-monomers in the DFT calculations is shown in Table S2. According to additional research, the Bi 6p state primarily affects BiOBr’s minimum conduction band (CBM), while the Br 4p and O 2p states account for the majority of its valence band maximum (VBM). In comparison, the S 3p state dominates the VBM of NiCo₂S₄, while the Ni 3d and Co 3d states mainly contribute to the CBM. While the Co 2p, B 2p, and C 2p states make up the majority of the VBM in BiOBr/NiCo₂S₄ composites, the Br 4p, S 3p, and O 2p states are mainly involved in the formation of the CBM involved. The BiOBr/NiCo₂S₄ composite had a significantly higher density of states (DOS) than the BiOBr and NiCo₂S₄ heterostructures, which increased the carrier concentration and improved the photocatalytic activity of the heterogeneous photocatalysts [55].

The high-resolution TEM images of the NiCo₂S₄/BiOBr-30 are shown in Figure 4. There were noticeable lattice fringes, with lattice spacings of 0.192 nm, 0.277 nm, and 0.281 nm corresponding to the (020), (110), and (012) lattice planes of the BiOBr, respectively. The (220) and (440) lattice planes of the NiCo₂S₄ were represented by lattice spacings of 0.332 nm and 0.166 nm, respectively. By demonstrating close contact and the development of a heterojunction structure between the two materials, these observations provided visual evidence of the interfacial structure between the NiCo₂S₄ and BiOBr. This further supported the formation of the NiCo₂S₄-BiOBr heterojunction, which corresponded to the XRD pattern results. In addition, the elements Ni, Co, S, Bi, O, and Br were present in the NiCo₂S₄/BiOBr-30, as shown in Figure 4d–i, according to the elemental distribution mapping images. NiCo₂S₄/BiOBr-30 was composed of the following elements: Ni is yellow, Co is orange, S is magenta, Bi is green, O is red, and Br is light blue. The NiCo₂S₄/BiOBr-30 composite sample exhibited a uniform distribution. These results indicate that the NiCo₂S₄/BiOBr-30 composite was successfully synthesized with good dispersion.

An X-ray photoelectron spectrometer was used to examine the surface elemental composition and valence state of the ideal photocatalytic composite sample NiCo₂S₄/BiOBr-30. As illustrated in Figure 5a, the XPS survey spectra indicate the existence of Ni, Co, S, Bi, O, Br, and C elements. Figure 5b displays the Ni 2p XPS spectra. Two spin-orbit doublets and two satellite peaks were observed. Ni²⁺ and Ni³⁺ existence in the NiCo₂S₄/BiOBr-30 was confirmed by the peaks at binding energies of 852.9 eV and 869.9 eV for Ni²⁺, while the peaks at binding energies of 855.7 eV and 873.3 eV were confirmed for Ni³⁺ [40,56]. In the Co 2p XPS spectra displayed in Figure 5c, two spin-orbit doublets and two satellite peaks were also observed: Co³⁺ was identified by peaks at binding energies of about 778.1 eV and 793.3 eV, whereas Co²⁺ was identified by peaks at 796.8 eV and 780.9 eV [57]. The S 2p XPS spectra is shown in Figure 5d, where the presence of metal–sulfur bonds is indicated by the peak at 161.2 eV, which is equivalent to S 2p^3/2. S²⁻ valence states were also present in the sample, as indicated by a peak at 162.4 eV [37,56]. Bi³⁺ was present in the NiCo₂S₄/BiOBr-30, as evidenced by the Bi 4f XPS spectra (Figure 5e), which also shows peaks fitted at 164.4 eV and 159.1 eV, which corresponded to Bi4f^5/2 and Bi4f^7/2, respectively [58]. Figure 5e displays the O 1s XPS spectra, where the peaks at 530.5 eV and 531.5 eV corresponded to O in the NiCo₂S₄/BiOBr-30 and surface-adsorbed components like -OH and H₂O, respectively. Figure 5f shows the Br 3d XPS spectra, where the peaks at 69.4 eV and 68.4 eV represented Br 3d^3/2 and Br 3d^5/2, respectively, indicating the presence of Br¹⁻ in NiCo₂S₄/BiOBr-30 [59]. It should also be noted that the peak at 65.7 eV was an interference peak from Ni 3p. These results not only confirmed the successful synthesis of the NiCo₂S₄/BiOBr-30 but also provide information about the composition and oxidation states of the corresponding elements, further helping to understand the heterogeneous NiCo₂S₄/BiOBr photocatalysts.

3.2. Photocatalytic Performances

To investigate the photocatalytic reduction activity of the NiCo₂S₄/BiOBr composite materials on Cr(VI), K₂Cr₂O₇ aqueous solution was employed as the simulated pollutant. Figure 6a illustrates the photocatalytic Cr(VI) reduction procedure. The photocatalytic reaction system included a 50 mL quartz glass reactor, a Xe lamp, and a magnetic stirrer. A total of 10 mg of the synthesized products was dispersed in 30 mL of 100 ppm Cr(VI) and ultrasonicated for 2 min. Then, the solution was continuously stirred for 30 min in the dark to reach the adsorption–desorption equilibrium before the photocatalytic reduction process. After this, the mixed solution was exposed under a 300 W Xe lamp with continuous stirring. Every 10 min, 1 mL of the suspension was sampled and filtered through a nylon membrane filter with 0.22 μm pores. The concentration of Cr(VI) was mainly determined by a colorimetric diphenyl-carbohydrazide (DPC) method by calculating the absorbance of the pink complex at 542 nm. The photocatalytic performances of the NiCo₂S₄/BiOBr heterogeneous composite with different mass ratios for the Cr(VI) removal under visible light irradiation are compared in Figure 6b,c. The NiCo₂S₄/BiOBr composite samples had an increased photocatalytic Cr(VI) reduction activity with the increase in the BiOBr mass ratio. More BiOBr in situ growth increased the specific surface area, which improved the separation of the photogenerated carriers and further stimulated the photocatalytic reduction reaction [60]. Among these composite samples, the NiCo₂S₄/BiOBr-30 showed the highest performance, with a 96.67% removal rate after 60 min of light irradiation. Furthermore, we compared the photocatalytic reduction performances of the NiCo₂S₄/BiOBr-30 with those of the pure NiCo₂S₄ and BiOBr; the absorption performance of the NiCo₂S₄/BiOBr-30 without light irradiation was also determined to understand the photocatalytic reduction performances of the heterogeneous composite. As shown in Figure 6d, the NiCo₂S₄/BiOBr-30 had a much higher Cr(VI) adsorption capacity than both the pure NiCo₂S₄ and pure BiOBr. The formation of the NiCo₂S₄/BiOBr heterostructure increased the specific surface area of the composite sample, which improved the adsorption performance of Cr(VI) [61]. It was confirmed that the adsorption–desorption equilibrium was reached after continuously stirring for 30 min in the dark. Furthermore, the improved adsorption performance of the NiCo₂S₄/BiOBr-30 composite samples enabled the treatment of higher Cr(VI) concentrations with less catalyst, which is beneficial for their use in the treatment of wastewater with high Cr(VI) concentrations.

Figure 6e displays the photocatalytic Cr(VI) reduction performances, where the NiCo₂S₄/BiOBr-30 showed much higher photocatalytic Cr(VI) reduction activities compared with the pure samples. The reaction rate constants (k) of the prepared samples for the photocatalytic reduction of Cr(VI) during the light phase were also calculated by assuming a first-order reaction kinetic model with the following formula [62]:

$$\ln ({\mathrm{C}}_0/{\mathrm{C}}_{\mathrm{t}})=k\mathrm{t}$$

(2)

where C_t and C₀ denote the Cr(VI) concentration at light times t min and 0 min, respectively. As illustrated in Figure 6f, using the light exposure time (t) as the horizontal coordinate and ln(C₀/C_t) as the vertical coordinate, the k and R² values were acquired through graphical fitting. The fitting of the NiCo₂S₄ and NiCo₂S₄/BiOBr composite samples yielded R² values that were nearly 1, suggesting that the catalysts’ photocatalytic reduction of Cr(VI) was entirely governed by pseudo-primary reaction kinetics. Additionally, the order of the k values was NiCo₂S₄/BiOBr-30 > NiCo₂S₄ > BiOBr. The NiCo₂S₄/BiOBr-30 composite samples had the highest photocatalytic reduction rate (0.0574 min⁻¹), which was almost 10 times that of BiOBr (0.0574 min⁻¹) and 3 times that of NiCo₂S₄ (0.171 min⁻¹). This improvement was mostly ascribed to the heterojunction construction, which raised the catalyst’s photocatalytic activity. Furthermore, the XPS measurements were carried out to analysis the NiCo₂S₄/BiOBr-30 before and after the reaction. A Cr XPS peak appeared after the reaction in Figure S3a, and the high-resolution XPS spectra of Cr 2p after reduction is shown in Figure S3b. Peaks at 576.8 eV, 577.9 eV, and 586.5 eV represent the binding energies of Cr(III), suggesting that the majority of the Cr(VI) in water had been reduced to Cr(III) after the photocatalytic reaction [63]. We also compared the performances with the mixed sample (7 mg NiCo₂S₄ + 3 mg BiOBr) and the commercial photocatalyst TiO₂(P25), as shown in Figure S4. The mixed system at the same mass ratio only resulted in a 59.5% reduction in Cr(VI), with a lower absorption and photocatalytic reduction efficiency compared with the NiCo₂S₄/BiOBr-30 composite. According to the above results, the adsorption performance and photocatalytic reduction activity of the composite samples were affected by varying the BiOBr coatings on the NiCo₂S₄ surface. The NiCo₂S₄/BiOBr-30 composite provided excellent adsorption and photocatalytic performances for Cr(VI) relative to those of the mixed sample and P25.

Because the pH can affect the catalyst material’s surface charge and the spatial arrangement of reactants, it is one of the most important experimental parameters that influence photocatalytic reactions. Even metal sulfides that are insoluble in neutral or alkaline environments may dissociate at low pH levels because of the presence of acids, causing sulfide ions to be released and dissolved. In order to examine the influence of initial pH on the photocatalytic reduction of Cr(VI) by the NiCo₂S₄/BiOBr-30 composite sample, the initial pH values were adjusted between 4 and 11 by 0.5 M HCl and 1 M NaOH solutions. The time profiles on the photocatalytic reduction of Cr(VI) by NiCo₂S₄/BiOBr-30 under different pH values was illustrated in Figure 7a,d under same reaction conditions (10 mg catalyst in 30 mL 100 ppm Cr(VI) concentration at room temperature). The NiCo₂S₄/BiOBr-30 composite’s dark adsorption and photocatalytic reduction activities toward Cr(VI) both dramatically increased when the pH decreased from 11 to 4, suggesting that slightly acidic conditions promote the photocatalytic reduction of Cr(VI). This was mostly because the catalyst surface became positively charged due to protonation at lower pH levels. Furthermore, Cr(VI) in aqueous solutions can be found as Cr₂O₄²⁻ (in alkaline conditions), Cr₂O₇²⁻, and HCr₂O₄⁻ (in acidic conditions) [64]. The adsorption of Cr(VI) in its anionic forms was facilitated by the catalyst’s more positive surface potential at lower pH levels. As a result, the photocatalytic reaction rate increased because there were more active sites available for the reaction. When the surface of NiCo₂S₄/BiOBr-30 was electrically neutral around pH = 7, there was less adsorption of Cr(VI) in its anionic forms, which lowered the reaction rate. The catalyst and Cr(VI) experienced electrostatic repulsion when the catalyst surface became negatively charged in an alkaline environment [65]. The lowest photocatalytic reduction rate was also the result of the reduced Cr(III) precipitating as Cr(OH)₃ and covering the catalyst surface under alkaline conditions. This further reduced the number of active sites available for the photocatalytic reaction. The NiCo₂S₄/BiOBr-30 composite reached photocatalytic reduction rates of more than 95% for Cr(VI) at pH 4 and 5. However, faster photocatalytic reaction rates could be attained without additional pH adjustment because the initial pH of the Cr(VI) solution was normally around 5.4 without a pH adjustment in our experiments. Overall, Cr(VI) was reduced to Cr(III) under different pH conditions as follows [32]:

Acidic condition:

$${Cr}_2{O}_7^{ 2-}+{14H}^{+}+{6e}^{-}={2Cr}^{3+}+{7H}_{2}O$$

(3)

$${HCrO}_4^{ -}+{7H}^{+}+{3e}^{-}={Cr}^{3+}+{4H}_{2}O$$

(4)

Alkaline condition:

$${CrO}_4^{ 2-}+{4H}_{2}O+{3e}^{-}={Cr\left(OH\right)}_3+{5OH}^{-}$$

(5)

The photocatalytic reduction of Cr(VI) by the NiCo₂S₄/BiOBr-30 composite sample could be influenced by co-existing ions because the contaminated water contained a variety of ions, as illustrated in Figure 7b,e. The photocatalytic reduction of Cr(VI) was not significantly affected by Cl⁻, SO₄²⁻, or NO₃⁻ ions, while the addition of HCO₃⁻ strongly inhibited the reduction of Cr(VI). The competitive adsorption between the anions and Cr(VI) was the main reason for this inhibitory effect. Figure 7c,f present the cations’ influences on the photocatalytic reduction of Cr(VI). It demonstrates that the addition of Cu²⁺ greatly increased the photocatalytic Cr(VI) reduction performance, mainly because Cu²⁺ could be more easily reduced to Cu⁺ using the photogenerated electrons, and Cu⁺ took part in the Cr(VI) reduction process [66]. Other cations, such as Mg²⁺, Ca²⁺, and K⁺, present in the solution could decrease the efficiency of the Cr(VI) reduction. These cations could both be competitively adsorbed (on the surface of the catalyst) with respect to Cr(VI) ions and reacted with photogenerated electrons. This reduced the probability that electrons were involved in the photocatalytic reduction of Cr(VI) and influenced the reduction process of Cr(VI).

To further investigate the photocatalytic reduction mechanism of the NiCo₂S₄/BiOBr composite samples, free radical scavenging experiments were carried out. The scavenger concentration was more than ten times higher than the Cr(VI) concentration using isopropanol (IPA), potassium bromate (KBrO₃), and methanol as scavengers for hydroxyl radicals, electrons, and holes, respectively. Nitrogen gas was also introduced into the reaction system to eliminate the dissolved oxygen from the solution, and thus, prevent the formation of superoxide radicals. From the photocatalytic reduction of Cr(VI) by the NiCo₂S₄/BiOBr, the results of the free radical scavenging experiments are shown in Figure 7g,h. It is obvious that the addition of methanol to the system slightly increased the photocatalytic reduction efficiency of Cr(VI) by the composite catalyst. The main reason for this improvement was that methanol reacted with the holes created by the lighting, which resulted in fewer holes. Furthermore, hole consumption promoted the participation of electrons in the reduction of Cr(VI) to Cr(III) by preventing the recombination of photogenerated electron–hole pairs. After the addition of KBrO₃, the reduction rate of Cr(VI) by the catalyst was 91% after 60 min of light irradiation, and the photocatalytic performance was reduced compared with the reduction rate without the addition of the capture agent (about 96%), suggesting that electrons were involved in the photocatalytic reduction process of Cr(VI). When nitrogen was pumped into the system, the reduction rate of Cr(VI) was significantly reduced by the NiCo₂S₄/BiOBr-30 composite sample. At the end of the 60 min photocatalytic process, the reduction rate of Cr(VI) by the catalyst was 84%, and the photocatalytic performance was significantly decreased, which was mainly because the dissolved oxygen content in the solution decreased with time since the continuous bubbling of nitrogen into the system prevented the formation of superoxide radicals. Therefore, superoxide radicals played an important role in the photocatalytic reduction of Cr(VI). In summary, electrons and superoxide radicals were the main active species in the photocatalytic reduction process of the Cr(VI) reaction process. The possible reaction process of the photocatalytic reduction of Cr(VI) in the presence of the NiCo₂S₄/BiOBr-30 composite sample is shown as follows:

$$Photocatalysts(Ni{Co}_2{S}_4/BiOBr)+hv\to e^{-}+h^{+}$$

(6)

$${Cr}_2{O}_7^{ 2-}+{14H}^{+}+{6e}^{-}\to {2Cr}^{3+}+{7H}_{2}O$$

(7)

$$H_{2}O+{2h}^{+}\to {1/2O}_2+{2H}^{+}$$

(8)

$${3•O}_2^{ -}+Cr\left(VI\right)\to Cr\left(III\right)+{3O}_2$$

(9)

Furthermore, the cycling stability of photocatalysts is an important consideration when assessing the feasibility of photocatalytic technology in wastewater treatment. Photocatalytic reactions were tested to investigate the cyclic stability of the NiCo₂S₄/BiOBr-30 composite sample. Before a 60 min light reaction, each cycle began with a 30 min dark adsorption procedure. The catalyst was filtered and dried at 70 °C for 12 h for the next cycle after rinsing three times with deionized water and once with 95% ethanol. Figure 7i shows the stability of the NiCo₂S₄/BiOBr-30 composite sample for the photocatalytic reduction of Cr(VI) over four cycles. According to the results, the photocatalytic reduction efficiency of the NiCo₂S₄/BiOBr-30 only slightly decreased with increased cycle number. In the photocatalytic Cr(VI) wastewater treatment, the NiCo₂S₄/BiOBr-30 composite sample showed some cycling stability, as evidenced by the fact that the removal efficiency of Cr(VI) remained higher than 75% after four cycles.

3.3. Enhanced Photocatalytic Cr(VI) Reduction Mechanism

The activity of photocatalysts is strongly influenced by the separation and transport of photogenerated carriers, the efficiency of photogenerated carrier separation will enable more active carriers to participate in the reaction, thus increasing the photocatalytic performance. Figure 8a shows the transient photocurrent responses of the NiCo₂S₄, BiOBr, and NiCo₂S₄/BiOBr composite samples. The BiOBr had the lowest photocurrent response between them, while the NiCo₂S₄/BiOBr-30 had the highest. This suggests that the NiCo₂S₄/BiOBr-30 heterojunction improved the separation and transport efficiency of photogenerated charge carriers compared with pure NiCo₂S₄ and BiOBr samples, which increased the photocatalytic activity [67,68]. The electrochemical impedance spectra of the samples are shown in Figure 8b, where the electron transfer impedance of the photocatalyst was indicated by the radius of the curves. The equivalent analog circuit diagram is shown in Figure 8b (inset), where R_s and R_ct represent the equivalent series and charge transfer resistances, respectively. The R_ct values for different samples are summarized in Table S1. Higher efficiency in the transfer of photogenerated carriers is indicated by a smaller surface radius, which is also reflected in a lower resistance in the photocatalyst [69,70]. The graph shows that the NiCo₂S₄/BiOBr-30 had the lowest charge transfer resistance at the interface, with the order of radius sizes being NiCo₂S₄/BiOBr-30 < NiCo₂S₄ < NiCo₂S₄/BiOBr-10 < NiCo₂S₄/BiOBr-50 < BiOBr. The enhanced photocatalytic reduction of Cr(VI) resulted from the efficient separation and transport of photogenerated electron–hole pairs. Consequently, NiCo₂S₄/BiOBr-30 exhibited maximum activity for the photocatalytic Cr(VI) reduction.

UV–Vis diffuse reflectance spectroscopy was used to determine the optical properties and bandgaps of the NiCo₂S₄, BiOBr, and NiCo₂S₄/BiOBr-30 composite. As shown in Figure 9a,b, the UV–visible diffuse reflection absorption spectra and the bandgap plots using the (αhυ)² versus (hυ) plotting method are presented. The pure NiCo₂S₄ had a visible light absorption property, while the edge in the absorption spectrum of the pure BiOBr had a blue shift compared with the NiCo₂S₄. The NiCo₂S₄/BiOBr-30 composite samples had significantly higher visible light absorption than the pure NiCo₂S₄ and BiOBr. Figure 9b allowed for the direct bandgap calculation by plotting (αhυ)² versus (hυ). The linear portion of the plot was then extrapolated along the tangent to a point where the Y-axis is zero and the intersection on the X-axis represents the bandgap value of the sample. Bandgap values of the pure NiCo₂S₄, NiCo₂S₄/BiOBr-30 composite, and pure BiOBr were 2.66 eV, 2.77 eV, and 2.86 eV, respectively. For the semiconductor photocatalysts, the band structure is crucial for the photocatalytic reactions and for determining whether a photocatalyst could be used for the photocatalytic reduction of Cr(VI). Mott–Schottky curves can be used to determine the flat band positions of semiconductors and reflect the relevant properties of semiconductor catalysts. The NiCo₂S₄ and BiOBr Mott–Schottky curves are shown in Figure 9d, where both were n-type semiconductors. According to the graphical method, the flat band potentials were −0.55 V and −0.46 V (versus the Ag/AgCl electrode), respectively. The conduction band potential differed by 0.2 V, which was more negative than the flat band potential for n-type semiconductors, which usually differs by 0.1–0.3 V [71]. In contrast to the normal hydrogen electrode (NHE), the CB positions of the NiCo₂S₄ and BiOBr were −0.55 V and −0.46 V, respectively. The calculated valence band potentials for the NiCo₂S₄ and BiOBr were 2.11 V and 2.4 V, respectively, based on the previously determined bandgap values.

According to the acquired bandgap and flat band potential data, the NiCo₂S₄ and BiOBr had matched energy level structures that could form a type-II heterojunction. This enabled effective carrier separation and transfer, which increased the photocatalytic activity of the heterogeneous NiCo₂S₄/BiOBr composite [72]. Figure 10 shows the proposed mechanism for the photocatalytic reduction of Cr(VI) by the NiCo₂S₄/BiOBr composite. When NiCo₂S₄ and BiOBr surfaces are exposed to light with energy equal to or greater than their bandgaps, electrons in the valence band (VB) will move to the conduction band (CB), producing photoexcited electrons in the CB and leaving matching holes in the VB. Since NiCo₂S₄ has a more negative CB and VB potential than BiOBr, the inherent electric field of the NiCo₂S₄/BiOBr heterostructure will improve the transfer of photogenerated carriers. Electrons in the NiCo₂S₄’s CB move to the BiOBr’s CB through the NiCo₂S₄/BiOBr-30 interface, while holes in the BiOBr’s VB move to the NiCo₂S₄’s VB, resulting in more separated carriers within the NiCo₂S₄/BiOBr-30 composite. The heterojunction’s construction also improves the catalyst’s ability to adsorb Cr(VI) ions. Separated photogenerated electrons with high reducibility can directly reduce the adsorbed Cr(VI) on the catalyst surface to Cr(III). Also, separated photogenerated electrons can reduce the oxygen adsorbed on the sample surface to create superoxide radicals; then, the superoxide radicals will reduce the Cr(VI) to Cr(III). These paths were proved by the free radical scavenging experiments.

4. Conclusions

Heterogeneous NiCo₂S₄/BiOBr photocatalysts with different ratios were prepared using a simple water bath heating technique, while NiCo₂S₄ nanoparticles were prepared using a two-step solvothermal method. The successful synthesis of NiCo₂S₄/BiOBr composite samples were confirmed by analytical methods, such as XRD, XPS, SEM, and TEM, which revealed useful heterostructures. When comparing the photocatalytic performance of the pure NiCo₂S₄, pure BiOBr, and NiCo₂S₄/BiOBr composite samples, it was found that the NiCo₂S₄/BiOBr-30 had significantly increased adsorption capacity and photocatalytic reduction activity over Cr(VI). The improved photocatalytic performance of the NiCo₂S₄/BiOBr-30 was mainly due to its increased specific surface area, which resulted in the stronger Cr(VI) adsorption capability and more active sites for the photocatalytic reduction of Cr(VI). The inherent electric field of the NiCo₂S₄/BiOBr heterostructure not only improved the separation and transfer of the photogenerated carriers within the composite but also improved the absorption of Cr(VI) ions and visible light. In this study, a visible-light-responsive heterostructure photocatalyst was developed, which could potentially be used in industrial and environmental remediation.